SUMMARY Short interfering RNA (siRNA) technology, capable of highly specific gene knockdown, has emerged as a promising molecular therapeutic tool in targeted cancer treatment. Nonetheless, only limited success has been achieved in the systematic administration of siRNA, due to a series of hurdles such as kidney filtration, uptake by phagocytes, aggregation with serum proteins and enzymatic degradation. Existing delivery systems (e.g. viral vectors or inorganic nanoparticles) all face significant challenges, such as high immunogenicity, toxicity, and limited tuning capacity in particle size/shape/functionality to achieve optimal efficacy. Versatile DNA nanoparticles (DNPs) offer a paradigm shift in delivery vehicles because of the unprecedented programmability of DNA nanotechnology. These DNPs offer unique advantages, such as excellent biocompatibility and minimal toxicity. However, the lack of comprehensive study to improve the delivery efficiency and specificity of DNPs has hindered their application in the therapeutic delivery of siRNA. We propose to perform a systematic study of DNPs as siRNA delivery vehicles, based on our main hypothesis that optimized physical (e.g. size and shape) and chemical (e.g. arrangement of targeting ligands) properties of DNPs can greatly enhance therapeutic siRNA delivery efficiency and specificity, and enable specific knockdown of target genes in vitro and in vivo for cancer. Utilizing our strong expertise in DNA nanotechnology, we will design and construct a library of DNPs in a high-throughput fashion, perform systematic investigation of how their properties affect the delivery efficiency, specificity, therapeutic efficacy both in vitro and in vivo with siRNA as the delivery cargo, and develop effective and safe DNP systems. The proposed experiments will be carried out in the following specific aims: (1) To determine the formulation of DNPs with enhanced delivery efficiency and specificity. Our initial work will focus on polyhedral structures of varied sizes and linear structures. Functional moieties (e.g. folic acid) will be incorporated into DNPs to improve their stability, targeting capability, and cell uptake efficiency. The experimental process will follow a design, test, optimization (redesign) cycle with five basic steps: (a) DNP design, (b) DNP construction, (c) morphology/stability characterization, (d) cell internalization, and (e) siRNA delivery and knockdown studies. (2) To examine the therapeutic efficacy, biodistribution and organ toxicity of Bcl-2-siRNA?conjugated DNP (DNP-siBcl-2). It is not known whether the surface chemistry of DNPs affects their efficacy, toxicity or biodistribution differently in vivo. We have chosen to knock down Bcl-2, which is one of the critical regulators of apoptosis. Here, we will examine folate receptor-targeted DNP-siBcl-2 (DNP-siBcl-2-FA) to determine therapeutic outcome, biodistribution and organ toxicity in vivo in a mouse model. Both cell and mouse models will help consolidate our findings to improve our understanding of siRNA delivery and the molecular mechanisms of action of DNP-siBcl-2.